Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 120 (2015) 902 – 907
EUROSENSORS 2015
Trafficsensnet Sensor Network for Measuring Emissions from Transportation P. Bryndaa*, J. Kopřivaa, M. Horáka a
Czech Technical University in Prague, Faculty of Transportation Sciences, Konviktská 20, 110 00 Prague, Czech Republic
Abstract
Trafficsensnet project is concerned with measuring levels of air pollution and noise related to road-based transport. New compact “master” sensor unit for measurements of CO, NO2, SO2, VOC, dust particles (PM2.5 ,PM10) and noise levels has been developed. Currently, 20 such units are being installed within the Spořilov district in Prague. Another unit has also been developed which may be used for gathering other measurements with low power requirements. In Spořilov, these “slave” units are currently used as weather stations to gather measurements of dispersion conditions which are important for predicting spreading patterns of pollution. © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license © 2015 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of EUROSENSORS 2015 Peer-review under responsibility of the organizing committee of EUROSENSORS
2015.
Keywords: Data collection, Environmental monitoring, Master-slave, Mobile nodes, Pollution measurement, Radio communications, Wearable sensors, Wireless application protocols, Wireless sensor networks
1. Introduction The main objective of the project is to provide current information about the environmental impact of roadway transportation to citizens and public authority. The aim of the project is to develop a sensor network capable of collecting environmental data pertaining to roadway traffic. Vehicular traffic is the most significant source of air
* Corresponding author. Tel.: +420-224359562 ; fax: +420-224229201. E-mail address:
[email protected]
1877-7058 © 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of EUROSENSORS 2015
doi:10.1016/j.proeng.2015.08.781
P. Brynda et al. / Procedia Engineering 120 (2015) 902 – 907
pollution in Prague [1]. A sensor network which densely covers selected area would provide location-specific data which could be used to create and verify propagation models of air pollution. Such models are used in planning and implementation of traffic engineering solutions. In the near future, traffic could be controlled by variable traffic signs based on current measurements of air quality. We are currently deploying 20 master sensor units – and 6 slave units (for gathering weather-related data) – as a part of the pilot project in Spořilov. The Spořilov district is surrounded by two roadways with high densities of vehicular traffic – they are used by 103 and 104 thousands vehicle per day respectively – and a third roadway which is used by 53 thousands vehicles per day [2]. About 14 thousand people live in this area of 2.05 km2. Before the beginning of the Trafficsensnet project, no sensors for measuring air pollution levels have been installed at Spořilov for extended periods of time. In Prague as a whole, there are currently only 20 other sensor units for measuring air quality [1].
Fig. 1. Deployment of master units in Spořilov
2. Architecture of the network The simplified architecture of the network consist of 4 layers. First of them is slave node layer which contain slave units communicating with master node via the IEEE 802.15.4 standard. Extended star topology is used for the connection between master and slave nodes. It is possible to connect up to 20 slaves to one master unit, which sends data via GSM network to a server. The master layer also includes mobile (wearable) sensors. These use Bluetooth low energy (BLE) standard to connect to a mobile phone which is used as a gateway for communication with the server. A proxy layer manages data from units and also allows import of data from other public sources (government institutions). On the other end, proxy layer provides data to our environmental DB and is also able to provide data to EuroGEOSS. The last – user – layer provides the environmental data to different types of users using a web interface. The users can also provide specific local information (local pollution, garbage, problem with noise) to the system through the website.
903
904
P. Brynda et al. / Procedia Engineering 120 (2015) 902 – 907
Fig. 2. Simplified architecture of the Trafficsensnet network
3. Units 3.1. Master unit The master unit was developed as a gateway unit for measurement of air pollution and noise. It is able to connect other (slave) units or to operate in a standalone mode. It uses GPRS/3G data for communication with the server and is capable of communication with slave units. Accurate time and location of the measurement are known thanks to an integrated GPS receiver. The GPS, in cooperation with an accelerometer, is used to secure the time and position synchronization. In the case of a fall or theft, master units send emergency packets with its position. Power outages are overcome by a built-in battery backup. Electrochemical sensors are used for emissions measurements. They have been chosen for their low power consumption, size and cost. The analysed air sample is drawn into the unit by a fan. It is possible to use temperature and humidity compensations. Dust is measured by an electro-optical sensor which is able to recognize dust particle larger than 1 μm. This type of measurement requires controlling the air flow which is ensured by control measurements on a sensitive thermistor. Photo ionization detector is used for measuring volatile organic compounds (VOC). An RMS A-weighted noise sensor was developed within the project for measuring acoustic noise. Table 1. Used sensors and parameters of the master unit Sensor
Type
Measuring range
Carbon Monoxide
Electrochemical
20 ppb – 500 ppm
Sulphur Dioxide
Electrochemical
15 ppb – 50 ppm
Nitrogen Dioxide
Electrochemical
15 ppb – 20 ppm
Volatile Organic Compounds
Photo ionization detector
1 ppb – 50 ppm
Dust
Electro optical
Dust particles > 1μm
Noise
RMS noise detector
50-100 dB(A) SPL
Dimension
(height x width x depth)
Value Weight Power Cover
98 x 120 x 120 mm 660 g
supply / consumption
7,5-36 V / 3 W IP 55
P. Brynda et al. / Procedia Engineering 120 (2015) 902 – 907
Fig. 3. Placement of the master unit and power source on a lampost in Prague district Spořilov
3.2. Slave unit Additional type of sensor platform has been created for measurements with low power requirements – a slave unit. Slave unit is a small, stand-alone platform which can be battery-powered and is therefore suitable for applications which have to be independent on external power sources. The platform is based on a microcontroller with a 2.4GHz transceiver (capable of wireless communication according to the IEEE 802.15.4 protocol), which is connected via the SPI protocol to an external memory which holds a unique device identifier and other unit-specific data, and a power source. Slaves may be equipped with different sensors (each unit is capable of holding up to 10 different sensors) and configured to gather a wide range of roadway traffic-related and environmental measurements according to the requirements of a specific application. In the experimental Trafficsensnet sensor network, slave units are used for determining dispersion conditions in a “weather station” configuration. Weather station slave units are equipped with internal temperature and humidity sensors and with a barometric MEMS sensor. Two external sensors are also used in this configuration – a wind speed pulse sensor and an analog wind direction sensor. The units are powered by two AA batteries. Currently, two additional types of slave units are being developed – a Doppler car detector and a magnetic car detector. These slave units are intended to provide the capability of determining current traffic flows and long-term traffic patterns to the Trafficsensnet network. 3.3. Mobile units Both master and slave units are intended for location-specific data gathering. To extend the coverage of the network into areas where no such units are mounted, two small, wearable mobile sensor units have been designed – “Bodywear 1” and “Bodywear 3”. Both Bodywear units hold a microcontroller with a BLE transceiver and either one photoionization VOC detector (Bodywear 1) or up to two different electrochemical sensors (Bodywear 3). Both units are powered by a pair of AAA batteries. Since these units lack the ability to transmit data directly to a server, a BLE link to a smartphone running a special application is used to send these transmissions. 3.4. Power sources For the optimal function, the densest possible network of the master units should be deployed. These units however need to be powered by an external power source. To this end, nodes of public lighting network may be used.
905
906
P. Brynda et al. / Procedia Engineering 120 (2015) 902 – 907
The master units need a continual power supply (the units however contain a battery backup, which enables them to function for approximately 6 hours). Streetlights are fully powered only at night, a specially designed power supply was therefore need to provide power even during the day hours when the streetlights are turned off. The UltraTiny power source was developed for this reason, it is capable of supplying the master units even in daylight hours when it is connected to a modified public lighting network. At night, the voltage in the streetlight network is 230V / 50Hz. During the daylight hours, the voltage is switched to lower value, which is not high enough to power the lights. The voltage used during the day is 24V / 50Hz. In other words, the public lighting switchboard contains two power sources. The first one, which is intended for use during the night, provides 230V / 50Hz and the second power source provides 24V / 50Hz for use during the day. Switching between power sources is accompanied by a time delay of about 1 minute. The UltraTiny power source has two voltage on modified street lamps. The first is active when street lighting lit. The second voltage during the day. It is designed to transform input voltages from 24V to 235V AC to an output voltage of 15V / 1A DC. The unit can be logically divided into 2 parts – an auxiliary power supply and a main power source. The power unit is controlled by a microprocessor which evaluates and controls the output voltage. 4. Preliminary results The Trafficsensnet network is deploying in Spořilov. Preliminary results revealed the need for detailed calibration of sensitivity, offset and temperature for the electrochemical sensors. Figure 4 shows an example of measured data (hourly average of concentration of CO) after the calibration in comparison with the certified reference measurements provided by Czech Hydrometeorological Institute. A good agreement can be seen. The time delay of the reference data is likely caused by an additional moving average applied on the data. For the noise measurement the difference from the certified measurements was about 1 dB(A) SPL. The network is able to provide timely data for information purposes. The next work is focused on precise calibration and verification of measured values.
Fig. 4. Comparison of hourly averages of carbon monoxid mesured by the master unit and a certified reference measurement provided by the Czech Hydrometeorological Institute
5. Conclusion The project is still in progress. Initial deployment of several units proved functionality of the architecture and communication protocols. Local residents and public sector considers the provided data as highly useful for their purposes. Further work will focus on continual deployment the sensory network in Spořilov and on the calibration of the sensors.
P. Brynda et al. / Procedia Engineering 120 (2015) 902 – 907
Acknowledgements This work was supported by the Grant Agency of the Czech Technical University in Prague, grant No. SGS15/226/OHK2/3T/16 and Technology Agency of the Czech Republic, grant No. TA02031405. References [1] City of Prague – Prague City Hall (2014): Prague Environment 2012 - Yearbook - report on state of the environment 55, 14-19. [2] Prague Transportation Yearbook 2013 (The Technical Administration of Roads of The City of Prague, Prague, 2014)
907
